Slashdot Mirror
Google Fiber Pauses Operations, CEO Leaves, and About 9 Percent of Staff Is Being Let Go (bloomberg.com)
The future of Google Fiber has been shaky ever since Google's parent company, Alphabet, was founded. The original plan was to expand Fiber's blazing fast internet service to more than 20 cities, with the goal of eventually delivering nationwide gigabit service. However, Alphabet hit the reset button on those plans Tuesday. Not only is Google Fiber CEO Craig Barratt leaving, but about 9 percent of staff is being let go. That translates to about 130 job losses, since the business has about 1,500 employees. Bloomberg reports: Barratt wrote in a blog post that the company is pulling back fiber-to-the-home service from eight different cities where it had announced plans. Those include major metropolitan areas such as Dallas, Los Angeles and Phoenix. Moving into big cities was a contentious point inside Google Fiber, according to one former executive. Leaders like Barratt and Dennis Kish, who runs Google Fiber day-to-day, pushed for the big expansion. Others pushed back because of the prohibitive cost of digging up streets to lay fiber-optic cables across some of America's busiest cities. "I suspect the sheer economics of broad scale access deployments finally became too much for them," said Jan Dawson, an analyst with Jackdaw Research. "Ultimately, most of the reasons Google got into this in the first place have either been achieved or been demonstrated to be unrealistic."
I live in a GF area and love it. There are three tiers, 5 Mbps for $0 (yes, free broadband), 100 Mbps for $70, and 1 Gbps for $90. They have been absolutely bulletproof, the speeds are for real when tested, and the online system and the way that it integrates with their WiFi router is awesome.
I have had multiple providers over the years, including Comcast and Verizon, and Google Fiber's product and service are easily better than the others.
If Google can't make this work, there may be no hope for anything better for a long time to come. I just hope I don't lose it here!
STOP . AMERICA . NOW
Well it may be true that no one can feasibly lay cable in a fiscally competitive way, but that isn't the only way to achieve a competitive marketplace. Redundant cable is really a pretty stupid way to do that - we could always just do what most countries do and implement unbundling access rules. In fact we have those rules on the books already, they were part of the telecommunications act of 1996, they just couldn't be exercised because of the FCC's stupid "third way" decision to classify ISPs as something other than telecommunications services. Now that they've reversed that decision, and once all of the lawsuits regarding that have been resolved, maybe we'll be able to have some competition among ISPs.
Per house, FTTH costs about $3000 on average to install. $25 per month over ten years. It's a long term investment, but it's far from being unreasonably expensive.
I don't think laying the cable is the hard or expensive part anymore. AT&T put fiber in my neighborhood last year, and what they do now is dig holes periodically and use a machine to just push the conduit right through the ground like an earthworm. I guess the holes are to make sure it's on course, I dunno. It's pretty cool and very fast.. they ran fiber past 200 houses and about 3 linear miles of street in less than a week with about 20 guys. I don't know how much the actual conduit and fiber cost, but figuring $20/hour per person it ended up being around $80 of labor per house for that stage. That's a lot less than I expected.
Great! Can you please elaborate? Particularly about equipment costs and stuff?
Cost is a complicated and depends on the situation but a very simple case probably would be somewhere between $5000-20000 per user for data cable for a simple run presuming there were multiple users along the route. High power lines can be far more expensive. Underground cable is somewhere between 4-8X as expensive to lay as overhead cable. The biggest costs are typically the civil engineering involved. Especially if you have to dig up or work around any existing infrastructure.
Equipment? Depends on what you are doing and where you are doing it. Ignoring the equipment to hook into existing infrastructure you're looking at trenching equipment, cable feeding equipment, and a variety of other goodies. You also have to watch for buried power, data, gas, water, and sewer lines which aren't always well documented. Dig by hand? Don't make me laugh. To do it right you have to lay the cable below the frost line in most cases which can be several feet deep in many places. I know code near me for a simple drop requires a minimum depth for cable TV cable of 18 inches.
It's pretty hard. I only have experience with a metropolitan area network but I suppose home fiber will have a similar cost. For a ~1 mile stretch consider the following:
a) You have to engineer the way the cables will go where they will terminate, what equipment you'll be using, where you'll be tapping off (fiber to each house or to a central unit).
b) You have to survey for existing cabling and make sure your installation doesn't encroach upon private property. This could be as easy as dialing a number and getting some plans or as expensive as having a ground radar and doing land surveys.
c) You usually have to notify and have permits for digging up. Sometimes the city will take it upon themselves, other times you have to do it. Cost of permits and notices to 100 houses
d) Install tubing, lay cables, 8 people, a backhoe and a dump truck will take a good week or two if all the surveys have been done correctly. Off course if you manage to hit a thing with the backhoe, you could be delayed for a day or two. You still have to pay the construction workers.
e) Fill the holes and repair side walks, lawns and streets you have broken open. Again, 4-6 people with equipment will take about a week doing that.
f) Fielding complaints, law suits, talking to locals why you didn't re-plant their lawn Kentucky Blue Grass etc
And then we haven't talked yet about material cost of lines, piping etc. Our estimates average about $500k/mile with a mix of overhead and underground cabling which isn't expensive since you can easily service 200 houses on a mile stretch. Underground is about 5 times as expensive as overhead. Obviously most of these costs are eventually fully funded by the tax payer (operators pretty much get paid for laying Internet lines) but it's a huge upfront cost and a player like Google won't benefit from grants since it's not really a local political player.
Last week we had a backhoe operator hit a line twice, the second time he literally ran off and went missing for 3 hours without notifying anyone. Obviously he doesn't work there anymore but it caused a delay of 2 days between fixing the line and finding a new licensed operator and bringing them on site.
Custom electronics and digital signage for your business: www.evcircuits.com
The Bergen Linux User group have actually done this for real (for all of 9 packets though), and it was also discussed here. Or course, that's now old hat as it doesn't implement QoS or IPv6oAC.
UNIX? They're not even circumcised! Savages!
If you read the actual blog, the only cities where "operations" are being paused are those classified as "potential" Fiber Cities. Existing Cities (KC, Austin, Nashville) and Upcoming Cities (Irvine, CA; Huntsville, AL; San Antonio) are still going to have Google Fiber.
Upgrading when you need to upgrade is stupid? Because communications technology gets more expensive over time?